Model save path: ./New_Models/bn_False_dataset_MNIST_epochs_200_lr_0.01_model_type_vgg11_rand_seed_265358_test_samples_None_train_samples_None_weight_decay_0.01.pth.tar
Training Set:
Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.09116753190755844
Inter Cos: 0.10967151820659637
Norm Quadratic Average: 23.567670822143555
Nearest Class Center Accuracy: 0.8079833333333334

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.12477564811706543
Inter Cos: 0.15512588620185852
Norm Quadratic Average: 39.25989532470703
Nearest Class Center Accuracy: 0.8011333333333334

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.16251002252101898
Inter Cos: 0.18391132354736328
Norm Quadratic Average: 43.173099517822266
Nearest Class Center Accuracy: 0.7879833333333334

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.19650430977344513
Inter Cos: 0.21486639976501465
Norm Quadratic Average: 46.547969818115234
Nearest Class Center Accuracy: 0.8231166666666667

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17336824536323547
Inter Cos: 0.24082395434379578
Norm Quadratic Average: 25.20828628540039
Nearest Class Center Accuracy: 0.8746166666666667

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2415972799062729
Inter Cos: 0.3077923357486725
Norm Quadratic Average: 15.411665916442871
Nearest Class Center Accuracy: 0.9142

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4185033440589905
Inter Cos: 0.3191254734992981
Norm Quadratic Average: 9.11082649230957
Nearest Class Center Accuracy: 0.9551333333333333

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6248084902763367
Inter Cos: 0.3997812569141388
Norm Quadratic Average: 8.817939758300781
Nearest Class Center Accuracy: 0.9817666666666667

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 1.816240906715393
Linear Weight Rank: 16
Intra Cos: 0.7363077998161316
Inter Cos: 0.33140748739242554
Norm Quadratic Average: 39.749122619628906
Nearest Class Center Accuracy: 0.9868833333333333

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 1.816370964050293
Linear Weight Rank: 2745
Intra Cos: 0.8263797760009766
Inter Cos: 0.32691001892089844
Norm Quadratic Average: 28.47516632080078
Nearest Class Center Accuracy: 0.9905833333333334

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.810764193534851
Linear Weight Rank: 9
Intra Cos: 0.8652638792991638
Inter Cos: 0.3277307450771332
Norm Quadratic Average: 20.066225051879883
Nearest Class Center Accuracy: 0.9918

Output Layer:
Intra Cos: 0.9052265882492065
Inter Cos: 0.3879172205924988
Norm Quadratic Average: 16.00307846069336
Nearest Class Center Accuracy: 0.99185

Test Set:
Average Loss: 0.03694288223385811
Accuracy: 0.9892
NC1 Within Class Collapse: nan
NC2 Equinorm: Features: 0.10398105531930923, Weights: 0.0457279309630394
NC2 Equiangle: Features: 0.2726126141018338, Weights: 0.2498401641845703
NC3 Self-Duality: 0.05143748223781586
NC4 NCC Mismatch: 0.005900000000000016

Layer 0: Conv2d(1, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.10219583660364151
Inter Cos: 0.12048852443695068
Norm Quadratic Average: 23.595197677612305
Nearest Class Center Accuracy: 0.8229

Layer 1: Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.13819169998168945
Inter Cos: 0.16982164978981018
Norm Quadratic Average: 39.34209442138672
Nearest Class Center Accuracy: 0.8181

Layer 2: Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.17789079248905182
Inter Cos: 0.20222090184688568
Norm Quadratic Average: 43.15596389770508
Nearest Class Center Accuracy: 0.8041

Layer 3: Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.21014273166656494
Inter Cos: 0.235158309340477
Norm Quadratic Average: 46.539306640625
Nearest Class Center Accuracy: 0.8401

Layer 4: Conv2d(256, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.18320606648921967
Inter Cos: 0.23825600743293762
Norm Quadratic Average: 25.153976440429688
Nearest Class Center Accuracy: 0.8916

Layer 5: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.2525542676448822
Inter Cos: 0.3324761390686035
Norm Quadratic Average: 15.39376449584961
Nearest Class Center Accuracy: 0.9242

Layer 6: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.4255698323249817
Inter Cos: 0.341042697429657
Norm Quadratic Average: 9.126602172851562
Nearest Class Center Accuracy: 0.9564

Layer 7: Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
Intra Cos: 0.6271106004714966
Inter Cos: 0.3835996389389038
Norm Quadratic Average: 8.88333511352539
Nearest Class Center Accuracy: 0.9779

Layer 8: Linear(in_features=25088, out_features=4096, bias=True)
Linear Weight Norm: 1.816240906715393
Linear Weight Rank: 16
Intra Cos: 0.740046501159668
Inter Cos: 0.3435681462287903
Norm Quadratic Average: 40.14655303955078
Nearest Class Center Accuracy: 0.9827

Layer 9: Linear(in_features=4096, out_features=4096, bias=True)
Linear Weight Norm: 1.816370964050293
Linear Weight Rank: 2745
Intra Cos: 0.8240528702735901
Inter Cos: 0.3454124629497528
Norm Quadratic Average: 28.816761016845703
Nearest Class Center Accuracy: 0.9863

Layer 10: Linear(in_features=4096, out_features=10, bias=True)
Linear Weight Norm: 1.810764193534851
Linear Weight Rank: 9
Intra Cos: 0.860931932926178
Inter Cos: 0.34416356682777405
Norm Quadratic Average: 20.318174362182617
Nearest Class Center Accuracy: 0.9872

Output Layer:
Intra Cos: 0.896486759185791
Inter Cos: 0.40314117074012756
Norm Quadratic Average: 16.21296501159668
Nearest Class Center Accuracy: 0.9875

